ralph-claude-code-autonomous-development

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 0.90). These URLs point to an unknown GitHub user and an opaque short domain and explicitly provide/encourage running raw shell scripts (install.sh / uninstall.sh via raw.githubusercontent.com and curl|bash), which is a high-risk pattern for distributing malware.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). This skill's SKILL.md explicitly instructs importing GitHub Issues ("Pattern 4: Import GitHub Issues" and ralph-enable --from github), which are public, user-generated third-party contents that the agent ingests and uses to drive its autonomous development actions, so external issue text could influence tool use and behavior.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill's install and CI instructions explicitly run remote code (git clone https://github.com/frankbria/ralph-claude-code.git followed by ./install.sh) and the uninstall uses curl -sL https://raw.githubusercontent.com/frankbria/ralph-claude-code/main/uninstall.sh | bash, which fetch and execute external scripts at runtime, so these URLs are runtime dependencies that execute remote code.